John Hare’s God’s Command, 8.2.2, “Jonathan Haidt”

Hare wishes to discuss Jonathan Haidt’s The Righteous Mind: Why Good People are Divided by Politics and Religion. The key is that Haidt defends the view we saw in Arnhart that evolution has given us a “groupish” attachment, one that is designed to make groups more effective at competing with other groups. Haidt goes on immediately to ask: “But is that really such a bad thing overall, given how shallow our care for strangers is in the first place? Might the world be a better place if we could greatly increase the care people get within their existing groups and nations while slightly decreasing the care they get from strangers in other groups and nations?”

His conclusion is that it would be nice to believe that we humans were designed to love everyone unconditionally. But rather unlikely from an evolutionary perspective. Parochial love—love within groups—amplified by similarity, a sense of shared fate, and the suppression of free riders may be the most we can accomplish. Religion is, he thinks, the crucial social practice that enables group formation. But should we really expect religion to turn people into unconditional altruists, ready to help strangers under any circumstances? Whatever Christ said about the Good Samaritan who helped an injured Jew, if religion is a group-level adaptation, then it should produce parochial altruism.

Our genes, on his view, under the prompting of religion give us parochial altruism, but not disinterested benevolence, or the kind of care that the Good Samaritan gave to the injured Jew. What is strikingly absent in Haidt’s account, however, is any exploration of the universalizing tendency of some religion. Religion is treated throughout as a “hive switch,” a group-level adaptation that gives us cohesion within the group together with competition against those outside it. But one theme of Hare’s book has been that we can find within the Abrahamic faiths not only tribal loyalty but divine commands that tell us to love or show mercy to the enemy and stranger and give us resources for doing so. The three arguments from the first chapter reveal an internal structure to this form of religion. If we are going to talk about the contribution of religion to morality, we need to take these features into account.

In 2001 Haidt published an influential article called “The Emotional Dog and its Rational Tail: A Social Intuitionist Approach to Moral Judgment.” In it he argued it was a mistake to follow the lead of Lawrence Kohlberg (and behind him, Kant, and behind Kant, Plato) in valorizing reason as the source of moral judgment. Rather, to use a different metaphor that’s central in The Righteous Mind, we should think of emotion as the elephant and reason as a rider who is controlled by the elephant. The contrast is with Plato in the Phaedrus (246a), who thinks of reason as the charioteer, controlling the two horses of ambition and passion. On Haidt’s picture there is nothing controlling emotion except other emotions.

He doesn’t contrast emotions with cognition he thinks emotions are in fact “filled with cognition,” and he moves to saying that the contrast is between two forms of cognition, which he now calls “intuition” and “reasoning.” But this is still confusing, because intuition has often been thought of as a kind of reasoning. Aristotle, for example, distinguished between nous (intuitive reason) and dianoia (discursive reason). One takes time, the other doesn’t, but both are rational. Haidt has misunderstood Plato here, thinking Plato’s telling us in the Republic that “passions are and ought only to be the servants of reason, to reverse Hume’s formulation,” so that philosophers are kings. But Plato does not say that philosophers are kings, or that passions are the servants of reason, but that they should be. Much of the Republic is a description of states or cities in which there is no rule by reason. The fact that we are actually ruled often by something non-rational does not show that Hume is right and Plato is wrong.

Haidt is also wrong about Kant. Hume’s victory over Kant is repeatedly trumpeted. But what is the operative picture of Kant here? He was “rather low on empathizing,” though not as low as Bentham, who probably had Asperger’s syndrome. And what’s the evidence for this? Haidt suggests that Kant provided an abstract rule, the Categorical Imperative, which is based in logic, and in particular in the law of non-contradiction. But Haidt does not seem to know the formula of the end-in-itself. According to this formulation of the Categorical Imperative, we have to share as far as possible the ends of all those we affect by our actions, and we have to make those ends our own ends. This requires us, Kant says, to sympathize. Haidt is trading in caricature.

Haidt’s view is that we should not think of God as giving us a command to universal morality, because there is no rational moral compass that could receive such a command, and no “inner scientist” trying to find the truth about how to live. Haidt has three kinds of evidence for the hypothesis that the intuitive dog wags the rational tail. The evidence comes from what he calls “dumbfounding” and “post hoc fabrication,” from psychopathy, and from the bias towards the self that is pervasive in moral justification.

To obtain the first kind of evidence Haidt tells his subjects stories that involve what he calls “harmless taboo violations,” and that he contrasts with “harm-based” stories like the one Kohlberg used to tell his subjects about Hans stealing a drug to save his wife. Here’s an example of a “harmless taboo violation”: a lab worker, a vegetarian, eats some human flesh (from a cadaver that was to be burned). Subjects presented with this vignette experienced a predictable flash of disgust. Only 13% said that what the person did was all right. But when asked to say what was wrong with what she did, the subjects seemed at a loss. Haidt says they seemed to flail around, throwing out reason after reason, and rarely changing their minds when it was shown their latest reason wasn’t relevant. People were making a moral judgment immediately and emotionally. Reasoning was merely the servant of the passions, and when the servant failed to find any good arguments, the master did not change his mind.

But even without an altogether clearly articulated vision of the good, we can still have such a vision that can shape the lives we try to lead. Suppose that what used to be pervasive in society was a justification of the prohibition of cannibalism or incest in terms of divine command: that these were against the order that God had established. But suppose this kind of justification has become less socially prevalent. We would expect people to become less articulate in their discursive reasoning. Dumbfounding may well be culturally relative, so that cultures that stress what Haidt calls “the ethic of divinity” are not dumbfounded by just the same stories. But from this cultural relativity it wouldn’t follow that the intuitions of people in those cultures were not tracking something actually bad, or that they didn’t have a conscience or rational moral compass whose job it is to do this tracking.

The data are important, because they show that we are less good at explicit discursive reasoning than we tend to think we are. But the data do not establish the conclusion that Haidt wants, namely, that the “rider’s job is to serve the elephant, not to act as a moral compass.” Again, we have here the slip between the descriptive and the normative.

Haidt uses the example of psychopathy to argue there’s no rational will or conscience whose job it is to act as moral compass. But how could this conclusion be established from the data of psychopathy? Even if there’s a genetic base for it, nothing follow about whether people without this condition have a faculty of reason that can guide them in more than strategic planning. Haidt has reduced reason to what Aristotle calls “cleverness,” which works out the means to any end presented. Aristotle says both practically wise and villainous people are called clever. But the evidence of our failures of practical wisdom does not show that we do not have the faculties that would make such wisdom possible, only that we do not exercise them reliably.

The third kind of evidence Haidt uses is from the bias towards the self that is pervasive in moral justification. He tries to show that reason is not fit to rule; it was designed to seek justification, not truth. What his data show, however, is something else, something he says in the very next sentence: People care a great deal more about appearance and reputation than about reality. There’s a key difference between these claims. The second is perfectly consistent with, and indeed supports, the Kantian view that we start off under the propensity to evil that overrides the equally innate but essential human predisposition to good. But the first denies this view, because it denies Kant’s account of the predisposition, which is that we are the sorts of creature who respond with a certain kind of feeling. Inside we often act more like a lawyer justifying ourselves than a scientist seeking the truth. Likewise when brain scans are performed on partisans when they hear about hypocrisy among their favored candidates. “The data came out strongly supporting Hume,” with emotional and intuitive processes running the show and only putting in a call to reasoning when its services are needed to justify a desired conclusion.

But, Hare responds, the fact that we pay attention to and delight disproportionately in thinking about what suits our own inclinations does not show that when we do so we are thinking properly, or that our reason is doing its “job.” Rather, it shows that we are not doing our job as rational animals at all well.

Hare concludes this subsection by saying a divine command theorist should take cognizance of the evidence of all three types (dumbfounding and psychopathy and bias), and should be chastened by it because of what it shows about our lack of intellectual virtue and some people’s lack of conscience altogether. But this should not make her abandon her theory. What she holds possible and what she holds obligatory depend on her theological premises, and what she thinks in particular about the three arguments presented in the first chapter. Evidence about our various forms of cognitive failure does not show that we do not have the ability to screen our initial inputs given the available assistance, or that universal morality is not an appropriate screen. If this is right, then this evidence does not show us that “parochial altruism is the most we can accomplish.”